In this paper, we developed a computing architecture and algorithms for supporting soft real-time task scheduling in a cloud computing environment through the dynamic provisioning of virtual machines. The architecture integrated three modified soft real-time task scheduling algorithms, namely, Earliest Deadline First, Earliest Deadline until Zero-Laxity, and Unfair Semi-Greedy. A deadline look-a-head module was incorporated into each of the algorithms to fire deadline exceptions and avoid the missing deadlines, and to maintain the system criticality. The results of the implementation of the proposed algorithms are presented in this paper in terms of the average deadline exceptions, the extra resources consumed by each algorithm in handling deadline exceptions, and the average response time. The results not only suggest the feasibility of the soft real-time scheduling of periodic real-time tasks in cloud computing but that the process can also be scaled up to handle the near-hard real-time task scheduling.
